Bug 93280 - [SNA] Inconsistency in line visibility when using different line-widths
Summary: [SNA] Inconsistency in line visibility when using different line-widths
Status: RESOLVED MOVED
Alias: None
Product: xorg
Classification: Unclassified
Component: Driver/intel (show other bugs)
Version: git
Hardware: x86-64 (AMD64) Linux (All)
: medium normal
Assignee: Chris Wilson
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-12-07 01:45 UTC by Romain Vavassori
Modified: 2019-11-27 13:41 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
archive containing program, patch and screenshots of the issue (7.23 KB, text/plain)
2015-12-07 01:45 UTC, Romain Vavassori
no flags Details
test program (1.61 KB, text/plain)
2015-12-07 01:48 UTC, Romain Vavassori
no flags Details
screenshot of the issue (4.46 KB, image/png)
2015-12-07 01:49 UTC, Romain Vavassori
no flags Details
screenshot of the patched issue (3.96 KB, image/png)
2015-12-07 01:50 UTC, Romain Vavassori
no flags Details
patch (on commit da9ad388) (487 bytes, patch)
2015-12-07 01:51 UTC, Romain Vavassori
no flags Details | Splinter Review

Description Romain Vavassori 2015-12-07 01:45:35 UTC
Created attachment 120376 [details]
archive containing program, patch and screenshots of the issue

There is an inconsistency in line visibility for lines of zero length
depending on the line-width attribute. This happens when using SNA instead of UXA with xf86-video-intel driver.

I have done a program to show this issue (see test.c and screenshots in attachment). The program draw a grid of sample lines with varying parameters:
- following X axis: increasing line length (0 to 2)
- following Y axis: increasing line-width (1 to 3)

Normally we expect to have the same behavior on each row of the grid (either visible lines or not) but in the case of zero length lines, we obtain a visible line only if line-width == 1.

From the docs [1] it is said that for a line-width >= 1, zero length lines should not be visible when drawn with CapButt attribute (as in our case).

I have put a patch in attachment solving the issue (patch.diff). Bug and patch have been tested with Debian Jessie official version (2.99.917) and also with git latest commit (da9ad388).

[1] http://tinf2.vub.ac.be/~dvermeir/manual/xlib/GC/manipulating.html

environment
-----------

uname -a:
Linux crunchbang 3.16.0-4-amd64 #1 SMP Debian 3.16.7-ckt11-1 (2015-05-24) x86_64 GNU/Linux

lspci -nn:
00:02.0 VGA compatible controller [0300]: Intel Corporation 4th Gen Core Processor Integrated Graphics Controller [8086:0416] (rev 06)

versions:
- xorg: Debian Jessie official releases
- xf86-video-intel: bug present in either Debian Jessie release (2.99.917) or latest git commit (da9ad388)

The driver used is i915 with SNA.

I can provide other logs if needed.
Comment 1 Romain Vavassori 2015-12-07 01:48:49 UTC
Created attachment 120377 [details]
test program
Comment 2 Romain Vavassori 2015-12-07 01:49:29 UTC
Created attachment 120378 [details]
screenshot of the issue
Comment 3 Romain Vavassori 2015-12-07 01:50:13 UTC
Created attachment 120379 [details]
screenshot of the patched issue
Comment 4 Romain Vavassori 2015-12-07 01:51:44 UTC
Created attachment 120380 [details] [review]
patch (on commit da9ad388)
Comment 5 Martin Peres 2019-11-27 13:41:16 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/xorg/driver/xf86-video-intel/issues/79.


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.